Investigating the cognitive phenotype of autism: a 3-year prospective longitudinal study

Background: Researchers have proposed that the core features of autism are caused by multiple independent cognitive atypicalities, including capabilities in piecemeal processing combined with difficulties in theory of mind and executive control. No study has examined the validity of this ‘multiple-deficits’ account with respect to developmental persistence, universality, and developmental (causal) relations across cognitive domains.

Objectives: The aims of this longitudinal study were threefold: (1) to establish whether this specific cognitive profile remains stable over time, (2) to determine the universality of this cognitive profile, and (3) to determine whether individual differences in early cognitive skills were predictive of later differences both within and across cognitive domains.

Methods: Sixty-nine children (38 children with autism and 31 typically developing children) involved in an earlier study on cognitive skills in autism were followed prospectively and reassessed 3 years later on visuospatial coherence tasks, simple and more advanced tests of theory of mind, and tests of executive function (specifically tapping planning ability and cognitive flexibility).

Results: At the group level, children with autism continued to show strengths in local processing and weaknesses in theory of mind and executive control, relative to typically developing children. At the individual level, however, this cognitive profile was not present in all children at either time point, especially at follow-up. Also, while individual differences within cognitive domains were reasonably stable over time, early theory of mind skills were found to be predictive of later executive function, but not vice versa.

Conclusions: These results of the group as a whole suggest that the cognitive profile in autism persists with development. Yet, the results on individual patterns of performance present a challenge for a multiple-deficits view of autism, and indicate that the developmental relationships between cognitive atypicalities are more complicated than presupposed.